Bypass grafting using autologous vein is a common and effective treatment for end-stage atherosclerosis, and over one million vein grafts are created annually in the United States. Although vein grafting is almost always initially successful, the development of neointimal hyperplasia eventually leads to stenosis and failure in 50-70percent of cases over the first five years, and remains the most significant obstacle to long-term patency. Neointimal hyperplasia is the direct result of uncontrolled vascular smooth muscle cell (VSMC) proliferation, which is stimulated in vein grafts by the profoundly increased pressure and flow of the arterial environment. As pharmacologic approaches to this problem have been uniformly disappointing, therapeutic strategies employing antiproliferative gene therapy have recently been suggested. Efficient adenoviral-mediated gene transfer into animal veins and vein grafts has been demonstrated in many laboratories, including the applicant's. The early programmed transgene expression is stable and leads to a modest reduction in vein graft neointimal thickening after four weeks. However, the lack of animal models that develop truly occlusive lesions (as opposed to simple thickening), as well as the uncertain toxicity and long-term efficacy of adenoviral-mediated gene transfer has hampered translation of this approach to the clinical setting. Furthermore, while efficient in thin animal veins, adenoviral- mediated gene transfer into human saphenous veins has been more problematic, due to the thickness of the smooth muscle cell layer and the natural barrier that the endothelium provides against large particle penetration. The purpose of this project is to develop better models of vein graft VSMC proliferation and to use these models to test the toxicity and tong-term efficacy of anti proliferative gene transfer using adenoviral and herpes simplex viral (HSV) vectors. The specific aims are: (1) to develop new models of vein graft VSMC proliferation in the experimental animal in vivo and using human veins in vitro, (2) to evaluate the toxicity and long-term efficacy of adenoviral-mediated antiproliferative gene transfer using these models, and (3) to evaluate the efficiency and toxicity of vascular gene transfer using HSV, a nonintegrating DNA virus with the potential for penetration and durable programmed expression in thick-walled human veins.